Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Neural networks for signal processing applications: ECG classification

N Mahalingam1, D Kumar

  • 1Department of Electronic and Communication Engineering, Royal Melbourne Institute of Technology, Australia.

Australasian Physical & Engineering Sciences in Medicine
|December 31, 1997
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Production of prolific microsheep by embryo transfer into large non-prolific sheep.

The Veterinary record·2006
Same author

Review article: immune suppression and colorectal cancer.

Alimentary pharmacology & therapeutics·2006
Same author

Achalasia cardia: A study of 113 patients managed with indigenous dilator.

Tropical gastroenterology : official journal of the Digestive Diseases Foundation·2006
Same author

Identification of a repetitive sequence belonging to a PPE gene of Mycobacterium tuberculosis and its use in diagnosis of tuberculosis.

Journal of medical microbiology·2006
Same author

Mutations in WNT7A cause a range of limb malformations, including Fuhrmann syndrome and Al-Awadi/Raas-Rothschild/Schinzel phocomelia syndrome.

American journal of human genetics·2006
Same author

Pandemic influenza and its implications for transplantation.

American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons·2006
Same journal

A comparison between EPSON V700 and EPSON V800 scanners for film dosimetry.

Australasian physical & engineering sciences in medicine·2020
Same journal

Nanodosimetric understanding to the dependence of the relationship between dose-averaged lineal energy on nanoscale and LET on ion species.

Australasian physical & engineering sciences in medicine·2020
Same journal

Schizophrenia diagnosis using innovative EEG feature-level fusion schemes.

Australasian physical & engineering sciences in medicine·2020
Same journal

Force decoding using local field potentials in primary motor cortex: PLS or Kalman filter regression?

Australasian physical & engineering sciences in medicine·2020
Same journal

EPSM 2019, Engineering and Physical Sciences in Medicine : 28-30 October 2019, Perth, Australia.

Australasian physical & engineering sciences in medicine·2020
Same journal

New name: Physical and Engineering Sciences in Medicine.

Australasian physical & engineering sciences in medicine·2020
See all related articles

This study optimized neural network performance for classifying QRS complexes using a modified Backpropagation algorithm. The enhanced algorithm achieved faster convergence, classifying complexes more efficiently.

Area of Science:

  • Cardiology
  • Artificial Intelligence
  • Machine Learning

Background:

  • Accurate classification of QRS complexes is crucial for electrocardiogram (ECG) analysis.
  • Traditional methods may face challenges in complex waveform recognition.
  • Neural networks offer a promising approach for automated ECG interpretation.

Purpose of the Study:

  • To enhance the performance of a Multi-Layer Perceptron (MLP) for QRS complex classification.
  • To investigate the impact of modified Backpropagation (BP) algorithm parameters on network convergence and accuracy.
  • To determine the optimal neural network architecture for this specific classification task.

Main Methods:

  • Utilized a Multi-Layer Perceptron (MLP) with a 20x20 bitmap representation of QRS complexes as input.

Related Experiment Videos

  • Experimented with varying numbers of hidden layers and neurons to assess convergence rates.
  • Modified the Backpropagation (BP) algorithm's weight change rules, incorporating momentum and learning rate variations.
  • Introduced a learning rate adaptation factor to improve convergence stability.

    • Main Results:

    • Larger neural networks generally converged more easily, but performance decreased beyond a certain size, indicating an optimal architecture.
    • The modified Backpropagation algorithm demonstrated superior performance compared to the original.
    • The optimized network converged in 9,000 learning cycles, significantly faster than the original algorithm's 14,000 cycles.

    Conclusions:

    • The modified Backpropagation algorithm significantly improves the efficiency and performance of MLP networks for QRS complex classification.
    • Finding an optimal neural network architecture is critical for maximizing classification performance.
    • This approach offers a more efficient and potentially more accurate method for automated ECG analysis.